wrote:

Before I start my rant, let me excuse myself to Len (who I highly
regard) for taking some issues to point on the actually-having-done-it
level.



"Gain" of a crystal radio depends on the bigness of the antenna.

If you are talking about a loop antenna on an AM [BC band]
radio, then it's a different story. The loop antenna on an
AM receiver is small/tiny/micro-stuff relative to the 200+
meters of AM BC wavelengths. The received signal VOLTAGE
is directly dependent on the number of turns in that loop and
the physical size of the loop.

Loop antennas do suck on a xtal set for that reason but "bigness" can
suck equally if not done right.



A humungous-long wire is going to supply the greatest amount of
POWER to a crystal receiver. POWER drives the headphones. But,
the amount of power coupled in involves IMPEDANCE and that, right
away, gets into a complicated mess of more electrical rules.

You know how you read all those old texts about using nice glass
insulators and keeping the wire away from anything? In everyday
practice you can do pretty well what you please with wire antennas and
powered radios and notice very little difference. Once you get into the
realm of truly hi-z installations those tree leaves brushing against the
wire become noticeable.

Simple crystal receivers want to keep impedances very high at
both input, middle, and output. ["crystal" or piezo-electric
headphones are the best for that, next best is the highest
impedance magnetic headphones (2000 Ohms or higher) you can get]

Acutally the best is a good matching xfmr and sound powered phones.
Those old 1920's 2k headsets are relatively deaf. But you did specify
"simple crystal receivers" and it that case it doesn't really matter.

For the typical parallel-tuned L-C input to a crystal set, the
inductor Q will make a difference. It must be as high as is
practical; Qs of 200 to 300 have been done. But, the Q of the
coil is dependent on a LOT of different factors which I noted
in the other message.

Q in excess of 1000 is readily achievable. 200-300 is a starting point
on a decent dx set.



How come the coils on many of the CR schematics


Mostly, that is just old-time tradition! :-) [I kid you not]

The formula for resonance is: F^2 = 1 / (39.478 * L * C)

.....snipping here....

The maximum to minimum variable capacitance ratio is equal to
the square of the maximum to minimum frequency tuning ratio
desired. That's about IT.

Most of these old ckts relied on a 17 or 21 plate cap which was in the
400-500 pf range. They would tune the entire band at full tapping. The
tapping allowed a guy to really nitpick his tuning for two
reasons...ease of fine tuning and higher Q at the best combination.

"Taps" on a coil can be to select different inductance values
for resonance with limited-range variable tuning capacitors.
Note: Back in the prehistory of radio, like around the 1920s,
variable capacitors were expensive and not so easy to get. A
few old-time crystal sets "tuned" via lots of coil taps using
a fixed parallel capacitor. I had a Philmore crystal radio kit
back in 1946 that did that. Very cheap kit. It worked, so-so.

I think you described the Philmore kit well. That was a toy.


My 4 cents worth.


-Bill

-ex- wrote:
Q in excess of 1000 is readily achievable. 200-300 is a starting point
on a decent dx set.

So what's a good inductance to DC resistance ratio for an inductor on
an xtal set? The one I wound is about 500uH, and I get a resistance of
3.2 Ohms.

Using the formula Q=2*pi*f*L/R, I get a Q for my coil of 981 (@1MHz).
But based on its performance, I KNOW it's not that good. I'm picking
up a couple stations at night, but just barely.

Dave

dave.harper wrote:


-ex- wrote:

Q in excess of 1000 is readily achievable. 200-300 is a starting point
on a decent dx set.


So what's a good inductance to DC resistance ratio for an inductor on
an xtal set? The one I wound is about 500uH, and I get a resistance of
3.2 Ohms.

Using the formula Q=2*pi*f*L/R, I get a Q for my coil of 981 (@1MHz).
But based on its performance, I KNOW it's not that good. I'm picking
up a couple stations at night, but just barely.

Dave

For BCB work the 'standard' is in the 220-240uh range for tuning with a
~365-400 pf cap. There's an (almost) infinite number of combinations
you can use if you want to split the band into segments which sometimes
has an advantage. But switches and tapped coils can also be Q-killers
once you get into the Q stratosphere.

In practice the coil Q is determined primarily by the form dielectric,
wire size, wire spacing, diameter/length ratio/neary coupling effects,
etc. R is far enough down the list that its generally not even
considered. When you do a DC measurement of coil R thats not
representative of the skin effects and true RF resistance, thats why the
textbook formula doesn't pan out.

If you want to make a fairly nice coil without getting into the expense
of litz, check out spider-web coils and rook coils. When done with say
16-18 ga wire, and diameters in the 4" range you can get a pretty nice
coil. With 166-strand litz (30-35c/ft) you'll note an improvement but
by that time its time to start thinking about a good hi-q ceramic
capacitor and circuit loading concerns.

The Rap-n-Tap forum is where to get some good info.
http://www.midnightscience.com/rapntap/ "Best coil" is a common topic!

-Bill

-ex- wrote:

In practice the coil Q is determined primarily by the form dielectric,
wire size, wire spacing, diameter/length ratio/neary coupling effects,
etc. R is far enough down the list that its generally not even
considered.

Ah, so the voltage drop across the coil (due to the small internal
resistance) and the close proximity of the wires give it some
capacitance? Does this affect performance or just screw add unwanted
capacitance?

Wouldn't adding space between wires cause some eddy currents and lower
the L of the coil?

Does wire coating make a difference regarding the dielectric? Or is it
another capacitance-altering effect?

Thanks again!
Dave

Comments interspersed, and staying with the BCB range scenario...


dave.harper wrote:

-ex- wrote:


In practice the coil Q is determined primarily by the form dielectric,
wire size, wire spacing, diameter/length ratio/neary coupling effects,
etc. R is far enough down the list that its generally not even
considered.


Ah, so the voltage drop across the coil (due to the small internal
resistance) and the close proximity of the wires give it some
capacitance? Does this affect performance or just screw add unwanted
capacitance?

The internal capacitance of the turns isn't enough to radically change
the basic LC resonance. Instead it tends to result more like dielectric
leakage

Wouldn't adding space between wires cause some eddy currents and lower
the L of the coil?

Again, not significantly in the BCB example. Take for instance, a 4"
diameter coil wound with #18 wire, however many turns it takes. Lets
say 60. Winding the coil close-spaced as opposed to about
one-wire-diameter spacing will require a few less turns (maybe 10%) to
get the same L. But the close spacing WILL result in lower Q once you
re-establish the same inductance. There can be more than one reason for
this...is it the winding spacing or the length/diameter ratio or more
dielectric loss that causes this? (Its certainly not the R). Its
impossible to say because you can't have one without the other!

Smaller coils, say toilet-paper tube size, don't exhibit this effect -
or at least not to the same degree. But there's a whole different
geometry there and its not optimum.

Nobody really knows exactly what goes on here other than trial-and-error
experiments to see how they behave.


Does wire coating make a difference regarding the dielectric? Or is it
another capacitance-altering effect?

Yes it does. One of the tests on a good high Q coil is to set it up on
a Q-meter then touch a piece of your coil-form material (or
wire-insulation) to the coil and see how it behaves on the Q-meter. It
shouldn't move. Lossy core material/insulation will cause a visible
effect with this test. I'll reiterate in case someone jumps in and
reads this without reading the earlier parts of the thread...you won't
see this happen with a low-q coil but as you get higher in Q it becomes
more and more evident. In fact, with a big solenoid coil and Q500 you
pretty much have to tie the sample material onto the end of a stick to
do this test because of hand effects. Not to be confused with resonance
detuning effects.

There's no good rule of thumb for insulated wire other than a test like
this. There's quite a bit of insulation material in 660-strand litz and
thats darn good wire. No way to make a comparison because BARE litz
can't exist! I don't think I've ever heard a comparison made between
say bare 16-18 wire vs enamelled. I tend to think any difference would
approach the 'too difficult to evaluate' range.

-Bill

-ex- wrote:
Comments interspersed, and staying with the BCB range scenario...


dave.harper wrote:

-ex- wrote:


In practice the coil Q is determined primarily by the form dielectric,
wire size, wire spacing, diameter/length ratio/neary coupling effects,
etc. R is far enough down the list that its generally not even
considered.


Ah, so the voltage drop across the coil (due to the small internal
resistance) and the close proximity of the wires give it some
capacitance? Does this affect performance or just screw add unwanted
capacitance?

The internal capacitance of the turns isn't enough to radically change
the basic LC resonance. Instead it tends to result more like dielectric
leakage

Wouldn't adding space between wires cause some eddy currents and lower
the L of the coil?

Again, not significantly in the BCB example. Take for instance, a 4"
diameter coil wound with #18 wire, however many turns it takes. Lets
say 60. Winding the coil close-spaced as opposed to about
one-wire-diameter spacing will require a few less turns (maybe 10%) to
get the same L. But the close spacing WILL result in lower Q once you
re-establish the same inductance. There can be more than one reason for
this...is it the winding spacing or the length/diameter ratio or more
dielectric loss that causes this? (Its certainly not the R). Its
impossible to say because you can't have one without the other!

Smaller coils, say toilet-paper tube size, don't exhibit this effect -
or at least not to the same degree. But there's a whole different
geometry there and its not optimum.

Nobody really knows exactly what goes on here other than trial-and-error
experiments to see how they behave.

Thanks again for the reply.

From what I've read here and elsewhere, I'm debating either making a
spider coil or a tight wound 4" cylinder ( about 1" length). Is there
any significant advantage to either?

I can see with a powered ferrite core how the spacing would make less
of a difference... but if tight winding results in a lower Q/other
effects, why space the windings for air-core, crystal radio coils,
period?

Thanks!
Dave

Huh? You wrote, "if tight winding results in a lower Q/other
effects, why space the windings for air-core, crystal radio coils,
period?" Do you not want a higher Q? Generally, people try for the
highest unloaded Q they can get, under some set of constraints.

Close spacing lowers the Q mainly because the current in the wire is no
longer radially symmetrical, if you look at a cross-section of the
(round) wire. That raises the RF resistance of the wire. For decent
(low-loss) form material, it's mainly the RF resistance of the wire
that determines the loss and therefore the Q. Generally, highest Q for
a given diameter and length is obtained by spacing the wire about two
wire diameters, center to center, at least for high frequency work. If
you want to use Litz wire, there's an optimum stranding...more, finer
strands are not necessarily better as you get to either lower or higher
frequencies. You should be able to find info on that, if you do some
searching.

There is such a thing as TOO HIGH a loaded Q. Let's say you start off
with a coil with unloaded Q of 500, and couple lightly to it with your
circuit (antenna and detector), so the loaded Q is 250. That means the
bandwidth at 3dB points, if you tune a station at 1MHz, is 4000Hz. If
you've tuned to the center of the station, your demodulated bandwidth
will be only 2kHz. Since the rolloff is gradual with a single-tuned
circuit, voice should be OK, but you'll be missing out on a lot of the
highs. (Mind you, it's not easy at all to get an unloaded Q of 500 at
1MHz!)

Using a very hack crystal radio--coil of about 3" diameter, antenna
just ten feet or so of wire, and an HP zero-bias schottky detector
diode--but into a low-noise audio amplifier--I've been able to listen
to standard broadcast stations a thousand miles away in the evening.
Biggest problem is getting rid of local stations...I'd use probably a
carefully designed three-resonator filter and a much better wire
antenna if I was serious about it.

Cheers,
Tom

dave.harper wrote:
-ex- wrote:



From what I've read here and elsewhere, I'm debating either making a
spider coil or a tight wound 4" cylinder ( about 1" length). Is there
any significant advantage to either?

They're going to be very similar in performance. If you do the solenoid
(cylinder), go with one wire spacing between turns instead of tight
wound. Thats been pretty well proven to give a bit better Q. And 16-18
ga wire is also in the 'best' range for both the spider-web and
solenoid. Next step up would be some serious litz wire.


I can see with a powered ferrite core how the spacing would make less
of a difference... but if tight winding results in a lower Q/other
effects, why space the windings for air-core, crystal radio coils,
period?

I'm not 100% sure I understand the question....On a
rook/basketweave/spider coil there's inherent spacing already. And a
solenoid coil will also do better that way...on larger coils. The way I
understand it is that there are several factors at work - primarily
interwinding capacitance and overall l/d ratio. In practice, as you
change one you also change the other. You can compensate one for the
other somewhat with a different diameter coil, different gauge wire, etc
but the general concensus is that 4"/16-18 wire/~1 wire diameter spacing
is pretty close to the best you can squeeze out of that class of coil.

HTH.

-Bill

From: dave.harper on Jul 17, 6:42 pm


-ex- wrote:
Q in excess of 1000 is readily achievable. 200-300 is a starting point
on a decent dx set.

So what's a good inductance to DC resistance ratio for an inductor on
an xtal set? The one I wound is about 500uH, and I get a resistance of
3.2 Ohms.

Using the formula Q=2*pi*f*L/R, I get a Q for my coil of 981 (@1MHz).
But based on its performance, I KNOW it's not that good. I'm picking
up a couple stations at night, but just barely.

The "R" in the Q formula is an equivalent resistance at frequency,
not just the DC resistance. That equivalent resistance is made
up of many things: winding form factor, wire size, and the DC
resistance to name the major factors.

Q alone won't determine sensitivity. Sensitivity, without some
accurate numbers such as transmitter power output, distance to
transmitter, antenna gain/loss, is going to be a very subjective
item. Even with them available the numbers can turn out to be
rather off when listened to.

A couple of years ago now, I wound a loop for 60 KHz (WWVB
reception) using #14 electrical wire. It was rather cheap
at Home Depot compared to enameled "magnet" wire for a 500
foot length. Inductance came out roughly according to formula
but the low DC resistance didn't do much for the Q. At 60 KHz
the Q was only about 68. :-) Dimensions were about 2 1/2 feet
diameter, circular, with an aluminum foil electrostatic shield
over the top of 57 turns. In retrospect I should have used many
more turns of smaller wire, such as #26 AWG, since signal
strength is proportional to the number of turns for the same
size loop.

It could have been the insulation on the electrical wire that
reduced the Q. Unknown. Would have to wind a similar one in
"magnet" wire to find out. It was measured for Q and inductance
without and with the foil electrostatic shield with no discernable
changes in Q, only slight in inductance. As it is, it works well
enough, is presently in the attic above the interior workshop.
[size dictated by trap door access to that part of attic]

Years and years ago I fooled around trying to make an AM BC
loop according to "expert instructions" from some magazine.
Spent a lot of time cutting the "blades" of the former to allow
zig-zag winding of some Litz wire someone gave me. Former was
3/32" phenolic laminate, cutting via a jig-saw. About 14 inches
wide by 6 inches high. Q measured out to only about 120 at mid-
band (using an old Booton Q Meter). Low enough distributed
capacity but not near the Q claimed in the article, supposedly
about 300. shrug Maybe ordinary cardboard would have worked
better as the former? :-)

If you have some RF source of known frequency at the AM BC band,
you can get a fair handle on the Q by using a high series
resistor between RF source and the L-C parallel-tuned circuit.
Observe the voltage across the L-C tank and de-tune the RF
source frequency to the 71% amplitude, note the two frequencies
on each side of resonance and take their difference. That's
the delta-F "Q bandwidth" that, when divided into peak resonance
frequency, will get you the approximate Q. The high resistance
source-to-tank should be around 100 KOhms or so (higher the
better) at 1 MHz to avoid introducing too much error. That
resistor forms a "quasi-constant-current" stimulus...not ideal
but good enough for an approximation when observing the RF
voltage across the L-C tank.

"Ours is not to reason why, ours is but to cut and try..." :-)

wrote:

Q alone won't determine sensitivity.

Very true. In fact the opposite is usually the case when it comes to
function. In a good hi-q xtal set one often has to sacrifice
sensitivity for selectivity...and vice-versa. Tighter coupling=more
sensitivity but less selectivity because of loading effects. If you
want sensitivity, go outdoors and hang wire or add sound-powered
headphones. If you want selectivity, tweak the ckt.

But still, by aiming for highest Q you at least have something to work
with when trading off.

Crystal radios are somewhat like a house of mirrors. Changes at one end
reflect at the other end of the ckt. The old razor-blade radio is an
example of everything being done the worst way but it still works. As
you improve on the scheme each component becomes more and more critical.

-Bill